The goal of this research is to clarify the pathogenesis of diabetic retinopathy and to provide a rational basis for developing improved means to inhibit the retinopathy. The proposed studies are based on our recent findings that inhibition of PARP dramatically inhibits the development of retinopathy in diabetes. In the present application, we will start from this novel finding to investigate two important gaps in our understanding of the pathogenesis of diabetic retinopathy. The first pertains to our lack of understanding of the sequence of biochemical abnormalities that ultimately leads to death of retinal capillary cells and other cells in diabetes. The second pertains to which cell type these abnormalities occur in. For many years, some investigators have suspected that circulating blood cells such as leukocytes play a role in capillary occlusion and the development of diabetic retinopathy, but this has not been rigorously tested. We will investigate the hypothesis that inhibition of PARP inhibits retinal pathology by inhibiting formation of nitric oxide, peroxynitrite, and prostaglandins. Moreover, by producing chimeric animals that lack PARP in their marrow-derived cells (leukocytes) but have it in retinal cells (or vice versa, have PARP in marrow-derived cells but lack it in retinal cells), we will test the hypothesis that PARP-mediated histopathology in the retina in diabetes is not caused merely by biochemical abnormalities within retinal cells themselves, but is due largely to sequelae of PARP activation in marrow-derived cells. Steps downstream of peroxynitrite and prostaglandin production responsible for hyperglycemia-mediated death of retinal capillary cells then will be investigated.